Systems and Methods for Background Noise Reduction in Magnetic Resonance Images

1. A method for visualizing a magnetic resonance (MR) image, comprising: segmenting the MR image into foreground which includes a region of anatomy of interest and background which is outside of the region of anatomy of interest; calculating an Otsu threshold based on the MR image; setting an intensity threshold to the Otsu threshold; applying the intensity threshold to the background and not applying the intensity threshold to the foreground of the MR image to produce a noise reduced MR image, wherein applying the intensity threshold to the background includes masking pixels or voxels of the background with intensity less than the intensity threshold; displaying the noise reduced MR image via a display device; receiving a user selected intensity threshold; applying the user selected intensity threshold to the background and not applying the user selected intensity threshold to the foreground to produce a second noise reduced MR image; and displaying the second noise reduced MR image via the display device.

2. The method of claim 1 , wherein segmenting the MR image into foreground and background comprises using a trained neural network to classify each pixel or voxel of the MR image as either foreground or background.

3. The method of claim 1 , further comprising: determining a maximum intensity of the background; and setting the intensity threshold to the maximum intensity of the background.

4. The method of claim 3 , further comprising: receiving a user selected intensity range; applying the user selected range to the noise reduced MR image; and displaying a subset of foreground within the user selected intensity range.

5. The method of claim 1 , wherein the MR image is one of a functional MR image, a diffusion weighted image, or a perfusion-weighted image.

6. A method for visualizing an MR image, the method comprising: segmenting the MR image into foreground which includes a region of anatomy of interest and background which is outside of the region of anatomy of interest, wherein the MR image is a diffusion weighted image; comparing an intensity of each pixel or voxel of the background against an intensity threshold; and responsive to the pixel intensity being less than the intensity threshold, setting a value of a corresponding pixel or voxel in a mask to zero; applying the mask to the MR image to produce a noise reduced MR image; displaying the noise reduced MR image via a display device; receiving a user selected intensity range; applying the user selected intensity range to the noise reduced MR image; and displaying a subset of foreground based on the user selected intensity range.

7. A magnetic resonance (MR) image processing system comprising: an image processing system; a threshold module; a display device coupled to the image processing system; a user input device coupled to the image processing system; and wherein the image processing system is configured to: segment an MR image into foreground which includes a region of anatomy of interest and background which is outside of the region of anatomy of interest; calculate an Otsu threshold based on the MR image via the threshold module; set an intensity threshold to the Otsu threshold; apply the intensity threshold to the background but not apply the intensity threshold to the foreground of the MR image to produce a noise reduced MR image; make the display device display the noise reduced MR image; receive a user selected intensity threshold via the user input device; apply the user selected intensity threshold to the background of the MR image and not apply the user selected intensity threshold to the foreground of the MR image to produce a second noise reduced MR image; and display the second noise reduced MR image via the display device.

8. The MR image processing system of claim 7 , wherein applying the intensity threshold to the background but not applying the intensity threshold to the foreground to produce the noise reduced MR image comprises: comparing an intensity of each pixel or voxel of the background against the intensity threshold; and responding to the intensity of the pixel or voxel being less than the intensity threshold by setting the intensity of the pixel or voxel to zero.

9. The MR image processing system of claim 7 , further comprising a storage device storing a trained neural network, wherein segmenting the MR image into foreground and background comprises using the trained neural network to classify each pixel or voxel of the MR image as either foreground or background.

10. The MR image processing system of claim 9 , wherein the trained neural network is a convolutional neural network.

11. The MR image processing system of claim 7 , further comprising a threshold module, wherein the threshold module is configured to: determine a maximum intensity of the background; and set the intensity threshold to the maximum intensity of the background.

12. The MR image processing system of claim 7 , wherein the image processing system is further configured to: receive a user selected intensity range via the user input device; apply the user selected intensity range to the noise reduced MR image; and display a subset of foreground within the user selected intensity range via the display device.

13. The MR image processing system of claim 7 , wherein the MR image is one of a functional MR image, a diffusion weighted image, or a perfusion-weighted image.